Beruflich Dokumente
Kultur Dokumente
Introduction
The Strategic Impact Inquiry is an attempt to better answer the critical question, are
CARE programs impacting the underlying causes of poverty and rights denial, and if so,
how? CARE has recognized gender inequality as a root cause of poverty across the
communities we serve, in particular through its impact on the capabilities of women.
Therefore, in this first SII on gender and power, we will focus on the question of CAREs
contribution to womens empowerment and gender equity.
Specifically, the SII will explore the following questions:
What evidence (pro and con) exists regarding the link between (a) CAREs
approaches and principles, (b) the advancement of gender equity and
empowerment? We will explore this through comparing the effectiveness of
different approaches CARE has pursued, and the implications of CAREs own
institutional form.
Investigating impact within the broad arena of gender and power is, by definition, a
multi-level, long-term challenge. It is, of course, fraught with challenges of measurement,
conceptual clarity, determining causes and effects, and teasing apart CAREs specific role
in contexts in which larger forces and a range of actors are also working to influence
gendered structures and relationships. In this context, the global research protocol is
offered to support the development of detailed, site-specific research designs. Its intent is
to offer a common, minimum core of guidance across all sites, including:
the basic purpose and guiding principles of this SII
key questions to serve as a unifying core across diverse research sites/questions;
critical dimensions along which we will explore evidence of change
research methodologies and approaches to ensure appropriate rigor
a draft timeframe of activities for Fiscal Year 2005 (through June 30, 2005)
The protocol, then, establishes a minimum, shared framework upon which site-specific
research teams should build, in order to facilitate comparisons across sites without
limiting each teams ability to specify the questions and methods in ways that best suit
their own needs. We emphasize that this global protocol does not seek to establish a
standard, detailed inquiry formula, or act as a logframe for this SII, and cannot
replace the need for broader or deeper site-specific protocols.
In support of this tri-fold purpose, certain core principles underlie this inquiry. The ethos
and approach of CAREs Strategic Impact Inquiry must be rights-based in itself, and so
must adhere to CAREs Programming Principles.2 A list of resulting principles that
should guide all SIIs is found in Annex 1. We will expand here on two that are paramount
in this particular SII on gender and power:
Joint participation of partners, the poor whom the program aims to serve, and
external researchers is a fundamental value that must be enacted. Knowledge
generated must be owned by the people whom CARE serves and not be produced in
an extractive manner. If at any point a choice must be made between research rigor
and participation of the poor and partners, participation must take precedence.
Respect for the physical and psychological safety of participants and informants
is paramount. Deep investigation into the operation of gendered structures of power
See Annex 2 for a brief overview of the wider array of inquiry elements being deployed within this years
SII on women and empowerment.
2
These are: Promote empowerment, Work with partners, Ensure accountability and Promote responsibility,
Address discrimination, Promote the non-violent resolution of conflicts, and Seek sustainable results. Fuller
treatment of these principles and the program standards that seek their enactment is available on request.
may put certain people (staff and participants) at risk of psychological or physical
violence in the short or long term. Every research team must review appropriate
guidelines, and be include an experienced researcher or programmer to ensure that
research methods do not put staff, partners or participants at risk. If at any point a
choice must be made between research rigor, data quality, and the safety and security
of participants, safety and security must take precedence.
III.
In this section we will outline certain choices explicitly made in narrowing from the
theme of gender and power to womens empowerment, and then lay out the structure
of questions that we propose to use to capture changes in womens empowerment. We
urge the reader to turn to Annex 3 for an overview of critical definitions and conceptual
frameworks that are key to this discussion and underpin the SII.
A.
First, this Impact Inquiry situates impact on gender and power in the context of CAREs
mission to end poverty and rights-denial. It does not explicitly undertake the broader
critique that feminist and subaltern studies advance of the patriarchal/hegemonic nature
of the development enterprise as such, although we believe that the values, principles,
and research methods we propose will open the inquiry to the issues these scholars and
practitioners have raised. For example, we remain open to the very real possibility that
research participants may put forward a very different vision of where they want their
empowerment to take them and call for local teams to allow indeed help women to
define crucial concepts like empowerment, equity, and equality themselves before
turning their eyes to whether CARE programs are having an effect on them. In addition,
we encourage local research teams to use any lens or set of analytical categories that they
would like to analyse CARE, its inner workings, and its industry context.
Second, the studys focus is on whether CAREs interventions are helping poor women,
in particular, to fulfill their needs and rights. While the relationship between womens
empowerment and impact on poverty levels more generally is an empirical question
worthy of study, it likely lies beyond the scope of this effort to demonstrate any causal
link between the two. Third, by the same token, while fully expecting a gendered lens to
explore how identities and opportunities are shifting for women and men, the focus on
women is in recognition that gender inequities often reflect womens subordination and
we wish to affirm womens importance in their own right. The issue here, we believe is
one of nuance and emphasis, but one that has specific implications for research design
and methods. We are making no claims of superior conceptualization here but,
rather, trying to make the actual research more manageable given the time and
resources available.
Third, the focus on womens empowerment represented by the research questions and
evidence categories shown on the next page is imagined as a fairly broad and inclusive
potential field of inquiry on womens empowerment. However, we fully expect that the
actual scope of research in any given site may well explore issues beyond those identified
on the one hand, and will very likely need to specify and narrow the range of issues
explored, on the other hand. At this stage, we encourage a conversation to identify the
crucial core in this year, and its expansion and movement in the future. We do hope that
by having an umbrella, global protocol that serves to frame, synthesize, and order
specific, local level findings that we can actually better organize, manage, and track that
longer-term, multi-level learning process that must become part and parcel of the
organizations ongoing culture of critical inquiry around issues of programmatic impact
on gender inequity and social exclusion. Once again it is important to remember that
this years SII does not purport to offer THE answer on CAREs impacts on
womens rights but a CONTRIBUTION to the global, organizational discussion and
dialogue on this very difficult question.
B.
This SII understands impact on womens empowerment to be reflected in three interconnecting aspects of social change. The first, driven by the actor-centered notion of
agency, is in the aspirations, resources, and achievements of women themselves. The
second is in the broader social structures that condition womens choices and chances.
And the third is in the character of the social relationships through which women
negotiate their needs and rights with other social actors. The first domain of inquiry, then,
is into the changes achieved in these three aspects.3
The second line of inquiry turns to CARE itself in order to understand our contribution to
change. The sub-questions here explore the approaches that the organization has used to
promote womens empowerment, and the relative effectiveness and implications of our
institutional strategies and forms.
Research Domain #1 (Impact Level):
What contributions (positive and negative) have CARE programs made, if any, to the
empowerment of women and the advancement of gender equity?
a.
(Agency question) What evidence is there that CAREs programs support the
expansion of womens capabilities to identify, pursue, and achieve their basic
needs and rights?
b.
c.
Annex 3 offers an overview of critical definitions and conceptual frameworks that underpin this SII, and
begins to illustrate the relationships that the research will explore between poverty, gender, power, and our
programmatic efforts.
What is the real mix of approaches (principles, models, strategies) that have
guided, and today guide, CAREs interventions? (Programming Principles
Review)
b.
c.
What types and mixes of interventions seem least effective, or counterproductive, in their impact on gender-equity and womens empowerment?
How prevalent are such practices in CARE programs?
d.
The following matrix offers key dimensions and illustrative sub-dimensions along which
previous research has measured and theorized womens empowerment, as well as a tool
for mapping how different perspectives might shed light on each dimension. A key
objective of the first gathering of the full SII team for FY05 will be to locate the research
questions in Bangladesh, Ecuador and Yemen with respect to this broad frame, and to
identify a common core around which teams can reasonably expect to generate useful
insights about impact.
Impact
Question:
What evidence
that programs
promote a
more
responsive and
equitable
enabling
environment?
Impact
Question
What evidence
that programs
promote more
interdependent
& accountable
relationships?
MDG/Rights
Defined globally and
in national law
RBA/HLS
Defined globally
The SII Team will review indicators drawn from good development practice, including
CAREs evolving rights-based livelihood security work, and identify a small number
that reliably convey key aspects of sustainable impact on womens empowerment to be
held constant across sites.
Question 1B
Poor women/men
Defined in context
The SII Team will review indicators drawn from global human rights commitments and
local legal frameworks, and identify a small number to be held constant across sites.
What evidence
that programs
support
expansion of
womens
abilities to
identify,
pursue, and
achieve basic
needs and
rights?
Evidence Categories
Dimensions of
Sub-Dimensions
Empowerment and
(13 proposed core across
Equity
sites are highlighted)
Agency-based
1. Self-image; self-esteem
2. Legal / rights awareness
Psychological
3. Information / skills
Legal
4. Educational attainment
Socio-cultural
5. Employment / control of labor
Political
6. Mobility in public space
Organizational
7. Decision influence in HH
Productive /
finance & child-rearing
Economic
8. Group membership / activism
Human / body
9. Material assets owned
10. Body health / integrity
Structural
11. Marriage/Kinship rules & roles
12. Inclusive & equitable notions
Cultural
of citizenship
(ethnicity, religion,
13.
Transparent information &
caste)
access to services
Legal / Judicial
14. Enforceability of rights, access
Market/Economic
to justice
Political
15. Market accessibility
Bureaucratic
(labor/credits/goods)
Organizational
16. Political representation
17. Share of state budgets
18. Density of civil society
representation
Relational
19. Consciousness of self / others
as inter-dependent
appreciation
20.
Negotiation/ Accommodation
flexibility
habits
cooperation
21. Alliance/Coalition habits
accountability
22. Pursuit / acceptance of
accountability
23. New social forms
Each site team should work with participants to develop key indicators of empowerment as
articulated by participant women and men themselves reflecting their own priorities and
conceptions, and informing our own.
Question 1A
Indicators / Evidence
Approaches
Question
Question 2A
What approaches have
guided, and today guide,
CAREs interventions?
Question 2B
What types/mix of
interventions most
promote gender-equity
and womens
empowerment? How
prevalent in programs?
Question 2C
What types/mix of
interventions are least
effective, or counterproductive, in impact on
gender-equity and
womens empowerment?
How prevalent in
programs?
Question 2D
Implications of institutional
form on degree/ nature of
impact achieved?
Institutional facilitating
factors? Key obstacles?
Effective
Principles
Strategies
Ineffective
Principles
Strategies
Orgl Implications
Structure
Staff mix
Culture
Governance
Civil society
positioning
Resources
IV.
Methodological Guidance
We all carry with us deep assumptions, values, and beliefs about what constitutes
evidence, proof, comparability, and plausible causal relationships. We may be divided by
very deep, philosophical and ethical disagreements on this issue or we may simply
wonder whether it is best, most effective, or most externally persuasive to demonstrate
causal relationships quantitatively or qualitatively. In this SII, we wish to dissuade CARE
from either/or reasoning on this important concern. Quantitative, experimental designs
with strict controls, leading to statistical reliability and validity for conclusions drawn by
the objective, outside researcher is right for some questions, some times, in some places.
Research deploying participative, action-learning focused, constructivist, actor-centered,
and emergent approaches, leading to qualitative forms of reliability and validity for
conclusions drawn jointly by external, CARE, CARE partner, and participant researchers
is right from some questions, some times, in some places. We strongly believe that in
trying to grapple with complex issues such as CAREs impact on gender inequity, we are
wisest to look to the strengths of each of these forms of scientific inquiry rather than
eliminate either from our palette of possibilities.
We need to be explicit and honest about the challenge that faces CARE as, over the
coming two or three years, it seeks to gather evidence of how it is affecting gendered
structures of power. One approach one weve heard suggested by some CARE leaders
would be to determine tangible, quantifiable indicators of womens empowerment and
ask all country offices to report on them on a yearly basis, perhaps through the existing
API process. It is possible that such transcultural, transnational, transhistorical indicators
exist but we are very timid, right now, about putting forth such a claim. As we all know,
gender, power, equity, and equality are all complex, many-faceted phenomena resisting
simple quantification. As a result, we believe it preferable, for now, to err on the side of
impact research that starts with womens own voices, interpretations, meanings,
indicators, and judgments rather than research that seeks to pigeonhole women into
frames imposed from the outside.
But we are not starting from scratch on these questions. The framework for
empowerment described in section III.B above has been drawn, already, from the
concrete experiences of women in scores of developing countries, across all regions of
the globe, as synthesized from hundreds of published studies. That frame should not be
applied unthinkingly, or mechanically, in Bangladesh, Yemen, or Ecuador, but we do wish
to use it as an intelligent starting point for research design and methods selection.
Few of the guiding research questions or evidence categories in the proposed
empowerment framework oblige the adoption or use of a particular methodology. The
most critical methodological decisions in this SII, therefore, will occur at the level of the
specific research sites Bangladesh, Yemen, and Ecuador and be a product of dialogue
between the external research team leader, CARE staff team members and, where and if
appropriate, partners and participants themselves. Methodological decisions will also
depend upon the specific research questions, project interventions, and analysis needs at
the three research sites.
Still, the FY05 SII seeks to draw reliable conclusions about our contribution to womens
empowerment and improving gender inequities. And as such, while methodological, data
gathering, and data analysis choices must be localized rather than universalized across
three very different kinds of projects, interventions, and operating contexts, there is a core
of methodological guidance that we wish to offer. The guidance is focused not on good
social science research design or methods in general4 e.g., principles and practices
around sampling, control groups, experimental/quasi-experimental/non-experimental
designs, cohesion between questions/methods/data/ analysis, data capture/treatment, etc.,
etc. but methodological recommendations or considerations specifically related to
researching womens empowerment. We divide these into the following categories:
A. Protection of participants
B. Getting at questions of meaning
C. Addressing power structures within the research process itself
D. Disaggregating the monolithic category women
E. Qualitative quantification (Participatory Numbers)
F. Looking at the larger context
G. Analysis
H. Methods, sources, and global evidence categories
Each section describes issues, obstacles, or challenges we believe may arise in all three
sites and suggests ways to overcome or transform them into opportunities. While not a
how-to or guide, the section offers some resources or paths forward that we can either
discuss more in depth in Cairo or pursue as local research unfolds.
A.
Protection of Participants
While all research involving people must adhere to basic ethical guidelines about the
protection from harm, investigations into womens empowerment can put women at a
particularly heightened risk of violence (by men who take issue with the nature of the
research and womens participation in it) and psychological trauma and retraumatization
(occasioned by women thinking about, remembering, or perhaps speaking to researchers
or each other about past violence, abuse, or present circumstances). The protection of
women in the context of the SII must be of paramount importance in the selection of
research designs, methods, approaches, sampling procedures, security of data, and forms
of analysis and reporting.
A consultant is currently helping us with guidelines (for use in orientation, training, data
collection, analysis and presentation) to ensure the safety and security of research
participants and teams and these will be distributed soon. We have also identified and
will bring to our first meeting some key references that you can consult to ensure that
your research does not put women at risk of physical or mental harm. No evidence is
worth putting women in your research area at risk. Period.
4
This research protocol assumes the contracting of well qualified social science researchers in Bangladesh,
Ecuador, and Yemen, as well as participation on the research team of CARE staff with strong backgrounds
and experience in the basics of good research. As a result, this section in no way pretends to offer a primer
or how-to on the ABCs of research design, methodologies, or data analysis.
B.
Getting at questions of meaning
We believe that to some extent, prior to trying to assess the strengths and weaknesses of
CARE programs, all three sites will be looking first at questions of interpretation and
meaning such as:
What do power and empowerment mean to women in the project zone? Does
it mean the same to all women? Are there patterns of difference and
similarity?
What does equity and equality mean to different kinds of women in the
project zone?
What does it mean to be socially excluded and is the experience of exclusion
the same for all women?
Starting with local definitions of power, empowerment, and so forth offers very firm
grounding in local realities and opens the possibility of having women themselves
develop the operational definitions of empowerment (as BOTH outcome and process)
against which you can assess and measure CAREs impact. These kinds of operational
definitions can then, actually, be transformed (with care!) into large scale, survey
questionnaire methods if you wish to/need to deploy such a method. A potential obstacle
or blind spot for researchers, however, is to utterly reject any universalist criteria
whatsoever for these concepts, thus risking in the worst of cases the reinforcement of
certain forms of exclusion, inequality, and inequity through romanticization of the
subaltern perspective.
Such challenges of getting at questions of meaning are not unique to studies of womens
empowerment, of course, but they are of tremendous importance if CARE wishes to
render itself accountable for its work in this arena. Certainly, some manner of direct
questioning of women themselves on these definitions is called for, either in individual or
small group interviews. But we need to consult the secondary, ethnographic, political
economic, and macrosociological literatures on these questions too. It is not necessarily
the case that individuals (men or women) are completely aware of how power, equity,
equality, and exclusion are at play in their own lives and we should be keen on this form
of triangulation and on helping women, as a result, make wider connections and build
deeper understanding of these dynamics in our research process itself. Other, less
obvious ways of getting at such questions of meaning include:
10
Ranking methods can also be quite helpful in getting at questions of meaning and
womens empowerment. For example, asking women to identify powerful or
empowered women in their social milieu, ranking them in terms of their power or
empowerment, then surfacing in the context of that exercise the categories and
indicators that help differentiate women from one another is sometimes very helpful
and can result in a quite nuanced understanding of forms of power and empowerment
in a relatively short time. Researchers can quickly saturate5 such categories in the
course of a relative handful of such interviews and triangulation can occur through
combining individual and group methods.
Finally, cognitive mapping (sometimes referred to as semantic mapping) can also help
to get at questions of social meaning. Such an exercise asks respondents to not just
define a particular concept, but to also do things such as define its opposite, define
something similar to it (and make clear the distinction between the two), and other
kinds of semantic/cognitive opposition or relationships. There are even ways of
quantifying such maps, although this is unlikely to be helpful in the context of our
work.
Category saturation is considered a good indicator of validity in qualitative research. It refers to the
phenomena with regards to questions of meaning, in any case to the stage when no new answers, no
new categories of evidence or response, are produced by additional research gathering techniques or efforts.
11
awareness and capacity building may affect your research design and implementation
plans.
Second, it behooves us to constantly remind ourselves to maintain a respectful and nonjudgmental stance in the research process, and avoid imposing our own assumptions or
values onto others. This is particularly important with regard to the deployment of the
global empowerment evidence categories described in section __ above: they should not
be thought of as normative bars which women, households, communities, or nations are
expected to cross, otherwise they are not empowered or not gender sensitive. Rather,
they should be thought of as initial guides for structuring an investigation process, a
starting place rather than a finishing line.
Third, we need to recognize that other forms of power are at work in our research
process, forms that may be even more important than gender. We have organizational
hierarchies that will permeate our research teams, for example. We face ideas about
power and knowledge in the communities where we work that constitute relationships
between development workers and community members, too. We may even face issues
of power over control, interpretation, and conclusions between our external, consulting
team leaders and CARE staff and participants. We have no magic, methodological
bullets on these difficult issues; rather, we encourage research teams to make these
relationships open topics of conversation and discussion throughout the research process.
D.
A common analytical error is to assume that all women are the same, that they all occupy
the same social position, are equally empowered or disempowered, are equally
advantaged or disadvantaged by the reigning culture and political economy. This is
clearly not the case. We need to try to understand a) which kinds of women are included
in our programs and which are not, and whether we ourselves our unknowingly excluding
some women, b) the spectrum of positive and negative impacts our programs are having
on different categories of women, and c) how we might shift or expand programs so as to
benefit the poorest and most marginalized women.
Our challenge will be to recognize and explore the diversity within communities, both in
the construction of a robust sample, and in the types of questions and lines of inquiry
pursued. Some key diversity factors to consider include gender, age, class, caste, religion,
ethnicity, family status, wealth, and livelihood strategies. The identification of different
categories of women need not be a complex, burdensome task, however. Staff
themselves frequently carry this information as tacit knowledge about their own societies
and cultures. Women can usually offer this information through interviews or
participatory group methods. Studying existing socioeconomic surveys or ethnographic
studies can help here, too. The challenge will be to identify the most strategically crucial
categories of women and sample across them.
12
E.
See, for example, R. Chambers and L. Mayoux Reversing the Paradigm: Quantification and Participatory
Methods EDIAIS Conference on New Directions in Impact Assessment for Development, U. of
Manchester, November 2003. There are a number of other papers from this conference that provide
guidance and case studies.
7
If the concept of participatory numbers strikes you as odd or, perhaps, unscientific, it really is little
different than such common, accepted, mainstream forms of qualitative quantification such as Likert scales,
IQ and psychometric test results, consumer confidence indices, etc. The main difference is that instead of
leaving scale, index, or test structure/content to Ph.D. experts, project participants construct them.
13
F.
Analysis
Every research method produces data that can be analyzed in a myriad of ways. This
brief research protocol is not a primer on data analysis techniques; rather we want
to simply underscore the importance of including women participants in CARE
programs in your analysis process.
As weve noted above, a fundamental value and principle of this research on the
empowerment of women in CAREs programs is that the research itself should be rights
14
based, should itself be empowering, and should itself lead to enhancing womens ability,
capacity, and confidence to analyze their own lives. We need, therefore, to ensure that
women are included in every step of the research process, have a voice and influence
over the questions we ask, the information we seek, the manner we seek it and, finally,
how we interpret it. This is not to say that poor womens analysis supercedes all other
perspectives but, instead, that their interpretations are equally important as others.
CARE staff are already familiar, for the most part, with a wide array of PRA, PLA, or
PAL techniques designed for just this purpose. The choice and sequencing of such
techniques is a local issue, one needing no treatment in this global research protocol.
However, we would like to call your attention to a couple of perhaps lesser known
techniques for embedding women in the entire research process as such participation is
perhaps essential for increasing womens abilities to analyze their situations and design
pathways forward.
One promising technique that has been used already in CAREs Zambia and Cambodia is
Peer Ethnographic Research. In this model still in a pilot phase of development by the
University of Swansea participants undergo a two-day training in interview and
observational techniques, then are given assignments to interview peers and observe their
communities. Peer ethnographers may be given tape recorders or even digital cameras to
make records of their work (and, in this way, even illiterate people can become impact
researchers). More information can be obtained at the Swansea U. web site.
A second promising technique for participatory data gathering and analysis is giving
women disposable cameras and asking them to go out and take pictures of things or
people that relate to the impact research terms of reference, evidence categories, key
questions, or indicators. After the film is processed the photographer can be either
interviewed alone or a larger group event can be organized in which many photographers
share their pictures, explain why they constitute data, and offer their analysis of the
data.
A third potentially empowering technique for participatory analysis involves bringing
larger, male and female, neighborhood, community, or even intercommunity groups
together in an action-planning event that is designed to identify concrete actions for
advancing womens positions. This may not work, of course, in settings where women
and men cannot speak or work together in such a way in a public space and great care
must be used in deciding that such an event is appropriate and will not result in physical,
emotional, or mental harm against the women who have participated in the research. The
interesting aspect of this technique is that rather than simply leave the research data and
analysis in the hands and minds of those closely affiliated with the CARE project, instead
the data and analysis drives a larger, wider, deeper consensus process that may even link
planned improvements with local political, religious, or economic bodies and thus give
them a higher profile and more importance.
H.
15
There is no single method, information source, or approach for analyzing any of the subdimensions of empowerment that the global research protocol identifies. All of the subdimensions are highly contextual and methods or data sources that are appropriate or
effective in Bangladesh may not be in Yemen or Ecuador. The table below is, therefore,
intended merely as an initial mental prod for your own creativity as you try to think about
your specific, local research questions. We offer them only as a starting point, a list of
commonly deployed methods and information sources in other studies of womens
empowerment that weve consulted over the past few months or, in some cases, particular
methods that weve seen deployed in only one or a small handful of studies but that seem
interesting.
16
Dimensions of
Empowerment
and Equity
A
G
E
N
C
Y
Sub-Dimensions
(13 proposed core across
sites are highlighted)
1. Self-image; self-esteem
2.
3.
Information / skills
4.
5.
Educational attainment
Employment / control of
labor
6.
7.
8.
Decision influence in HH
finance & child-rearing
Group membership / activism
9.
S
T
R
U
C
T
U
R
E
17
R
E
L
A
T
I
O
N
A
L
V. Implementation Timetable
The following timetable is proposed to guide our site-specific and global planning for FY
2005. It is understood that this timetable will be subject to the capacities and capabilities
across our proposed research sites however, with a strong wish to reach some useful
and interesting working results by the end of June 2005, we will nonetheless push for
an aggressive schedule for implementation.
Calendar
OctoberDecember
2004
2.
3.
4.
5.
JanuaryMarch
6.
2005
7.
18
April-June
2005
9.
9.
We propose that the research undertaken during the current fiscal year should only be the
beginning of a necessary process of deepening our understanding of the ways in which
CAREs work is contributing to gender equity and womens empowerment. Returning to
the three-fold purpose of this inquiry (increased accountability, impact and learning), we
realize that, internally, the findings across these initial research sites will only begin to
discover the impacts and strategies across CAREs global program. And we hope that
externally, that the findings and hypotheses generated this year will position us to deepen
our learning and accountability in collaboration with a wider movement of practitioners
and researchers who are committed to these goals.
19
Probe issues of gender but, more generally, power, marginalization, and exclusion, no
matter the theme;
Bring participants into the research process as more than simple informants;
Adopt learning process approaches that seek to build skills in CARE, partner, and
participant groups around impact assessment;
Offer mutual benefits to COs, regions, and the global organization; and
Annex 2
The FY05 Strategic Impact Inquirys Global.Methodological Quiver
20
21
One purpose of strategic impact inquiries in CARE is to permit the organization to speak
more empirically, authoritatively, and rigorously about the impact it is having on the
underlying causes of poverty. A criticism of much NGO research on gender, power and
empowerment is the lack of a clear and robust conceptual framework against which the
evidence can be analyzed.8 To this end, the following notes frame some of the key
concepts, models, and frameworks that lay the foundations for how CARE proposes to
explore its impact on the empowerment of women, and the advancement of gender
equity.
Modeling Poverty, and its Causes
CAREs mission is to fight poverty, and as such, our approach to questions of gender and
power is largely grounded in the impact that these forces have on the ability of people to
break the poverty cycle. CAREs concept of poverty has been deeply shaped by its
humanitarian and development assistance work, with its focus on the barriers people face
to meeting their material human needs on a sustainable basis. However, it has also been
deeply impressed over time by the voices of poor women and men, who have pointed out
to our own staff and to outside researchers the non-material dimensions on which they are
deprived and which they consider to be central features of poverty stigma and
discrimination, invisibility, hopelessness.9
CARE does not aim to develop a general theory of poverty, nor a particular definition.
However, to help our staff to think about these different dimensions of poverty and its
underlying causes, there have been significant efforts in recent years to bring them into a
more coherent conceptual framework. The basic framework that guides CAREs thinking
about poverty and the choices that confront the poor as they pursue their needs and rights
has been the Household Livelihood Security framework (HLS, see CARE UK
framework graphic, next page).
See, for example, Malhotra et al (2002), It should be noted that these criteria led to the effective
exclusion of most of the reports emerging from NGO programmatic efforts at empowering women. Many
of these reports lack the conceptual and empirical rigor we felt was necessary for inclusion in the current
review. (p.22).
9
See, for example, Narayan et al (2000), Voices of the Poor study prepared for World Banks 2000 WDR.
22
(ENABLING ENVIRONMENT)
Resources/ Services/ Opportunities
Water, Sanitation, Health, Shelter,
Land, Security/Safety, Education,
Work/ employment, Markets
Social opportunities
Control of resources/opportunities
by structures and processes (including executive,
judicial or legal institutions, organisations,
customs, policies, and legislation), e.g. control of
water by authorities How
inclusive/participatory/transparent are these?
Do they exclude some people?
Barriers to access:
resulting from ones identity e.g. race, nationality, sex, caste,
language, religion, political
opinion, origin / status or the
abuse of power
Assets
Household
Social
Physical
Human
Financial
Natural
Political
Knowledge
Pressure
Natural or people
made disasters,
aggression by
people in power,
conflict, illness,
seasonalities
(HUMAN CONDITION)
(SOCIAL POSITION)
23
As described in the excerpt below, HLS is an attempt to model, and thereby more
strategically support, the choices that poor people make in order to meet their basic needs
for a life with dignity, in the face of shocks and constraints.
A livelihoods approach emphases the capabilities, assets, and activities required for a
means of living. The most frequent definition used is that: A livelihood comprises the
capabilities, assets (stores, resources, claims and access) and activities required for a
means of living; a livelihood is sustainable which can cope with and recover from stress
and shocks, maintain or enhance its capabilities and assets, and provide sustainable
livelihood opportunities for the next generation (Chambers and Conway 1992). A
sustainable livelihood therefore requires the presence of sufficient capabilities and assets
to achieve the resilience required to cope with stresses and overcome shocks. The
measurement of livelihood status is usually conducted through assessing asset levels and
livelihood outcomes, the latter usually seen in material terms. Improvements in either
assets and/ or outcomes would be seen as indicators of livelihood improvement.
Conversely, decline in either areas would be a measure of livelihood deterioration.
An adequate analysis of livelihoods needs to be holistic in nature, encompassing an
understanding of context, social differentiation between households, and social
disaggregation within households, particularly with respect to gender and generation.
Nevertheless, whilst livelihoods approaches are adequate for ensuring effective
identification of poor and vulnerable groups, they place less emphasis on an analysis of
power, and the measures required to achieve greater social equity.10
This framework has been enriched through the years by CAREs work on rights, gender,
governance and other manifestations of power relations, breaking down the unitary
notions of a welfare-maximizing household, and generating much richer understanding of
the material, political, and social dynamics underpinning household options and
behaviors. At the same time, these new insights also generated a certain degree of
conceptual confusion among competing lenses.
One attempt to re-organize our thinking has led recently to the development of a second
conceptual model CARE calls the Unifying Framework. The Unifying Framework is a
very broad causal hypothesis about the production and reproduction of poverty in the
countries where CARE works. It seeks to focus CARE beyond the immediate and
intermediate causes of poverty that occupy much of our attention to date, and organizes
the underlying forces that shape the options and behaviors of the poor into three key
categories of causes those related to basic material or human conditions; to identity,
influence, and social positions; and to the structure of laws, norms, and institutions that
constitute the enabling environment.
The Unifying framework is intentionally meant to be flexible, open, and fluid; its major
function is to spark local conversation and analysis. Rather than try to introduce into
CARE any new theory of poverty, or to force a particular definition of poverty, the
Strategic Impact Inquiry will instead adopt the Unifying Framework as its own
theoretical model, understanding this to be an alternate and current representation of the
dynamic HLS framework.
10
SOCIAL POSITIONS
(Improving Social
Equity)
Equity:
gender, ethnicity,
caste, faith, age
Social
Inclusion
Mutual Respect
For Rights &
Responsibilities
Equitable
Distribution
Capital &
Assets
Voice &
Organizational
Capacity
HUMAN CONDITIONS
(Increasing Opportunity)
Productivity,
Livelihoods,
& Income
Human
Capabilities
Accumulation
Capital &
Assets
Risk &
Vulnerability
Management
Access Resources,
Markets &
Social Services
ENABLING
ENVIRONMENT
(Improving Governance)
Open &
Equitable
Government
Systems
Sound
Environmental
Stewardship
Social
Assistance
Protection
Civil Society
Participation
Strong & Fair
Environment for
Economic Growth
Fair Domestic
& International
Regulatory Framework
Conflict
Mitigation
26
27
While gendered structures of power can be said to produce general rules of dominance
and subordination across cultures, and result in exclusions from the resources and
opportunities needed to secure a sustainable livelihood, these examples demand constant
questioning of any orthodoxies regarding gender and its link to power and poverty. An
exploration of gender and power, then, requires a close and deeply sensitive attention to
the strategies deployed by individual men and women, and their collectives, in
negotiating the gendered power structures, and modulating behaviors of compliance and
defiance.
Power and Empowerment
A key breakthrough in CAREs evolving understanding of the underlying causes of
poverty has been the explicit recognition of power as the currency of material and
social well-being. Theories of power are not in short supply. Yet much of the
professional development literature particularly so-called gray literature that
discusses power lacks any explicit theory of power. Globally, CARE has only begun
work that might lead to the adoption of a common theorization and definition of power.12
The SII will build upon that work and try to move it forward.
Power has been simply and instrumentally defined by gender activists within the
development industry as the ability to get what you need, keep what you have, and
influence others in order to meet your interests. The simple but intuitive appeal of this
definition is grounded in the more sophisticated treatment given by Sen and subsequent
theorists to the notion of capabilities, and empowerment as the process of expansion of
capabilities.13 The capabilities approach, in turn rests critically on the concept of agency,
the active exercise of choice in the face of power relations and structures.14
The theory of power that CARE has begun working with and which underpins the work
of most theorists mentioned so far is strongly rooted in Anthony Giddens structuration
theory (1984).15 Central to Giddens theory are several simple notions:
1. Agency. The individual person is central in society: people are deeply
knowledgeable for Giddens. The can step back and assess the context in which
they act, they can even talk about the large-scale structures that might act as
constraints to freedom of action, and they can decide to take action in conformity
with such structural constraints or in contradiction to them. The individual no
matter his/her social identity always has agency. Large-scale structures are
12
See Elisa Martnez, Notes on Understanding and Measuring Empowerment, May 23, 2004, available
upon request from author martinez@care.org. This paper was presented at the May 2004 meeting of the
DME Cadre in Egypt.
13
See Sen, Kabeer, Nussbaum and others (get cites)
14
After Kabeer, Gender Mainstreaming in Poverty Eradication and the MDGs IDRC 2003.
15
Anthony Giddens, Elements of the Theory of Structuration, In The Constitution of Society: Outline of
a Theory of Structuration, 1-40. Berkeley and Los Angeles: University of California Press, 1984.
Martinez work marries gender theory to the World Banks effort to better measure empowerment. While
the Bank does not cite Giddens, his work underpins scholarly work referred to by the Bank.
28
human creations for Giddens, and people can therefore change them. Nobody is
powerless; nobody is all-powerful.
2. Structure. Agents (i.e., individuals) produce and reproduce routines,
conventions, relationships and taken-for-granted behavior. Over time these
become givens and we enact them largely without thinking why or how. Such
patterned, reproduced, and constantly reinforced relationships are what Giddens
considers social structure. All social structures come to seem objective, as out
there in society, as outside of our control. They imply deep, unspoken rules
which are deeply implicated in the reproduction of social relations, rules that
often lie hidden behind formalistic rules such as law. bureaucracy, politesse,
language, etc. Structure accomplishes three crucially important social goals: It
establishes agreed-upon significations (meanings), accepted forms of domination
(who has power over what or whom), and agreed criteria for legitimizing the
social order.
3. Subjectivity. Social structure (and the vast array of rules, norms, conventions,
etc. that become second nature or normal for societal actors) is imprinted in
the minds of social actors in two ways: Practical consciousness and discursive
consciousness. For Giddens, practical consciousness is not normally accessible
to agents: it is unconscious. Discursive consciousness is precisely what people
can articulate about their own actions and motivations. The discrepancy between
practical and discursive consciousness is critical to structures of power in any
societyand critical to effective research on the impact of CAREs programs on
gendered structures of power.
4. Resources. Power is inescapable in social life. As long as there are groups of
people power is at work. For Giddens, power is at work over resources. Thus,
any social grouping is organized to have resources flow and accumulate in certain
ways. And two categories of resources are critical for Giddens: Allocative and
authoritative. Allocative resources are those capabilities that command control
over objects, goods, or material life. Authoritative resources refers to control over
people.
The graphic to the right
schematicizes Giddens
model of the production and
reproduction of social life.
Structure(s)
Rules and resources, or sets of
transformation relations, organized as
properties of social systems.
Out of time and space
29
organize collective action. Another consequence of Giddens thinking is that power is not
necessarily a zero-sum game: just because one social actor has power does not mean that
others do not, nor does the accumulation or loss of power by a social actor automatically
mean that some other social actor gained or lost the same amount. And finally, power
itself is a socially constructed category, one that does not necessarily have the same social
signification in, say, Harlem, New York, as it does in Kandahar, Afghanistan.
Finally, Giddens theory of structuration points us to the different modes in which
power operates the fact that it is simultaneously present and acting in a number of
dimensions. Wolf (1990) has called these four modes of power,16 , and we have adapted
that framework to illustrate its link with forms/modes of power commonly discussed in
gender literature17:
Personal: Power within, Power To: An individuals ability to know, pursue, and, in
some cases, achieve their interests. Based on self-images, other-images, skills, resources,
and motivations
Interpersonal: An individuals ability to influence other agents and structures around
her/him, in order to achieve their interests can be cooperative (power with) or
controlling (power over). This latter takes important implications when applied to the
structural domain, in visible, hidden, and invisible forms:
Visible: derives from the formal/public forms, rules, and processes governing the
interpersonal process. EG membership in collectives, electoral laws, budgets.
Hidden: determines which agents/agendas become part of the interpersonal
process and the ability to control (often behind the scenes) the settings in which
agents interact.
Invisible: define, through processes of acculturation, the very field of the
possible, the reasonable or the logical. Examples include kinship in some
societies, capitalism, religion, science, and education. This kind of power
comprises and maintains the macro political economy and serves to define the
possible field of action of others.18
The graphic on the next page offers not so much a theory but a conceptual model based
on these adaptations of Giddens theorization for power that will be used in this SII. It
is used as the basis of the core evidence categories, indicators, data sources, data
gathering methods, and data analysis methods for the SII.
This model expresses how power dynamics drive interaction among material, social, and
environmental causes of poverty found in CAREs Unifying Framework. It also places
more centrally the creation and contestation of resources that is core to the HLS
framework. It reveals the underlying causes to be historical, relational, and relative
rather than static and absolute. The model also suggests the cohesion among CARE
Internationals six program principles. Each of the principles aims to generate programs
that are effective in addressing the web of causation shown in the model.
16
30
We propose that this model of power can be held constant for research purposes
across CAREs future inquiries into how RBA and a focus on poverty alleviation
articulates with power in the societies where CARE works. The model also provides
CARE researchers with the broad evidence categories they should incorporate into
research designs.
AGENCY
(and historical social
processes
and
practice)
Discursive
Consciousness
Practical
Consciousness
STRUCTURE
(and norms, assumptions,
unwritten rules,
enactment)
Authoritative
Modes of Power
Personal (power within, to)
Self- and other-images
Skills, capabilities, and resources
31
Empowerment has been theorized from many perspectives including those founded in
a more zero-sum notion of power and those that take a more expansive notion of
power. For the purpose of this study, we will focus on those discussions of empowerment
that take place within a feminist, gendered perspective. Empowerment is defined broadly
as the expansion of assets and capabilities of poor people to participate in, negotiate
with, influence, control, and hold accountable the institutions that affect their lives.19
Notable in this definition is the recognition of empowerment as a process of building
capability (and not simply the material outcomes visible in CAREs impact frameworks
to date), and of the importance of structure as represented by the institutions affecting
peoples lives.
This broad conception can be further grounded in a feminist theory as the expansion in
peoples ability to make strategic life choices in a context where this ability was
previously denied to them.20 This definition is notable in its focus on choice, which
Kabeers feminist perspective defines as comprising three critical elements: agency
(power within/to), operationalized in reference to resources (power to/over), and made
visible in its resulting beneficial/valued achievements. And finally, agency is exercised,
in this conception of empowerment, in opposition to a prior condition of subordination in
important (strategic) arenas of life. Strategic interests, in gender and development
theory, differ from practical gender needs, in that they go beyond the basic
functions/capacities which allow people to fulfill the gender roles assigned to them, and
aim to open new gendered spaces of ideology, action and opportunity. In this sense,
empowerment is importantly tied to impact on the structural underpinnings of womens
subordinate status and well-being.
Linking the Concepts to the Research Questions:
The schema on the following page is less a conceptual model than a visual way of
organizing the relationship between the various key concepts and the key questions
driving this inquiry. Its usefulness, if any, is simply in clarifying the many arenas in
which questioning will have to collect information and probe assumptions:
in the gendered nature of poverty outcomes and their causes in any given
community
in the nature and gender dynamics of household livelihood strategies
in the dimensions and degrees of empowerment that do or dont arise
in the relationship between CAREs approaches and organizational form and any
changes wrought in womens conditions, positions, and environments.
19
32
GENDER
SUSTAINABLE LIVELIHOODS
(a framework for understanding how people
act to achieve needs and rights, in the arena of
poverty and its underlying causes)
THE IMPACT
QUESTION:
What contributions
to the empowerment
of women and the
advancement of
gender equity
Evidence grounded in
agency womens
behaviors and
capabilities?
Evidence grounded in
structural changes
social norms and
forms?
EMPOWERMENT
(a partisan proposition and tool for
breaking the poverty cycle, based on a
marriage of power and gender theory
PROGRAMMING APPROACHES
AND PRINCIPLES (NEW/OLD)
(an organizational hypothesis, of the behaviors
and approaches that best support poor peoples
efforts to shift power and gender dynamics, and
overcome poverty)
What combinations of approaches have most
effectively led to meeting needs and fulfilling rights?
THE
APPROACHES
QUESTION:
What difference does
it make what
approach CARE
takes?
What approaches have
we been / are we
using?
Which seem effective?
Ineffective?
Organizational
implications?
33